Publication Type: Poster
Abstract: Students’ growth in statistical reasoning can be enhanced by providing opportunities to reason about theoretical probabilities using empirical data. This study examined how students in two Advanced Placement Statistics classes made statistical inferences by investigative analysis of empirical data using a computer simulation. This study concentrated on sample sizes used, evidence provided to justify inferences, and estimation of theoretical probability.

Publication Type: Session Presentation
Abstract: This panel applies the Sociocultural Theory to CFL learning in four different ways: a meta-study of current CFL research, discovering the rationale behind learner code-switching, factors affecting heritage learners’ language maintenance, issues facing life-long learners of Chinese. This panel calls for more CFL research from the Sociocultural perspective.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This article seeks to contribute to the debate surrounding the resilience of the welfare state by analysing pension reforms in Belgium and Sweden, in the 1990s. Reforming public pensions has been considered among the least popular undertakings a government can attempt since it affects a large portion of the electorate. By comparing the cases of Belgium and Sweden, this paper argues that the occupational divide in pension politics, clearly underlined by the work of Baldwin (1990), remains entrenched when seeking to introduce pension reforms regardless of the type of welfare regime.

Publication Type: Proceeding
Abstract: The hermeneutic circle provides a straight forward answer to the question when to stop researching, and, as a corollary, when to stop interviewing. The premise for accepting this answer lies in recognizing that any scientific research must start with theory, as only a theoretical framework allows for the separation of realms for systematic inquiry. First, we have to define what we are interested in, then our theory will tell us what variables we have to look for and how we suppose they relate to the phenomenon in question. In other words, the hypotheses we formulate allow us to determine what is relevant to our inquiry. Once we have separated a realm for our inquiry, we can start the process of gathering data, where speech acts are considered part of the data to be collected. In addition to collection speech acts, we must contextualize this data with other information relevant to the speaker(s) so that we can reach an understanding of her lifeworld and situate her speech. This is achieved by going for and back between the specific and the general, the concrete speech act and the political, historical, psychological, and in general institutional context in which the speaker and the speech is embedded. This conceptions leads us to gather empirical data up to the point when each single new information “makes sense,” i.e. it complements the logical structure of the lifeworld we are exploring. Each new interview must relate to and ultimately confirm what we already have found out, in a positive of negative way, and little by little we construct a contextualized understanding of the single speech act in question, which allows us to interpret each new piece information and locate it within the horizon of meanings that constitute the context or lifeworld of the speech and the speaker and the realm we have separated for our inquiry.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: In “Skating on Thin Ice,” Frohlich and Oppenheimer (2006) present a phenomenon they observed in public goods experiments that is rarely discussed in the literature: individual contributions to the public good are often inconsistent over time, appearing to fluctuate between two distinct contribution levels. They conjecture that a utility function with self-interested and other-regarding motives, modified by context-dependent preferences, could explain these events. In a related paper, Wendel and Oppenheimer (2007) develop a theory of context-dependent preferences and present results from an agent-based simulation testing two specifications; as is customary, we skip over the development of the theory and present the only these final outcomes.

I build on Wendel and Oppenheimer (2007) by analyzing the role of our agent-based model in developing a theory of context-dependent preferences. The goal is two fold. First, I examine the specific, step-by-step development of that theory. Second, I weave this experience together with a review of the literature to analyze of the benefits and pitfalls of agent-based models in theory development. While there is a substantial body of literature on the capabilities of agent-based models (e.g. Miller and Page 2007), relatively little has been written on theory development beyond issues of model validation.

While agent-based models have the capability to address exciting new questions in political science and beyond, there is unfortunately significant room to improve upon the conceptualization of theories underlying agent-based models, the testing of these theories, and their later refinement (e.g., Leombruni and Richiardi 2005). I note some of these challenges. I also find that, as agent-based modelers, we can learn from recent writings in the field and especially from decades of research in simulation modeling to provide more rigorous, and more compelling, theoretical arguments.